Publish In |
International Journal of Advances in Science, Engineering and Technology(IJASEAT)-IJASEAT |
Journal Home Volume Issue |
||||||||
Issue |
Volume-9,Issue-1 ( Jan, 2021 ) | |||||||||
Paper Title |
Traffic Safety Management using Spatial Analysis and Clustering Methods | |||||||||
Author Name |
Mohd. Minhajuddin Aquil, Mir Iqbal Faheem | |||||||||
Affilition |
Research Scholar, School of Engineering, Career Point University, Kota, India Prof. & Head & Vice Principal, Civil Engineering Department, Deccan College of Engineering and Technology, Dar-us-Salam, Hyderabad, India | |||||||||
Pages |
6-11 | |||||||||
Abstract |
The efficiency of traffic safety is influenced by the traffic accidents which results in severe lapse in identification of hotspots. This research focuses on identification and priority ranking of traffic accident hotspots using Getis-Ord Gi* and Nearest Neighborhood Hierarchy (NNH) methods which is solved using QGIS and CrimeStat softwares. QGIS tool calculates the Getis-Ord Gi* statistic for each feature in a dataset and the resultant separates traffic accident hotspots clusters of high values from clusters of low values spatially, whereasCrimeStat software detects road traffic accident hotspot that focuses on groups of data that are spatially close. Getis-OrdGi* method can visually represent the traffic accident hotspots and the Hotspot Analysis (NNH) provides the statistical significance of the hotspots in terms of NNH statistics. In this study, the road network of city of Hyderabad in the State of Telangana is selected as the study area which is demonstrated using the application of spatial clustering analysis tools such as Getis-Ord Gi* and NNH methods. In this study, a cumulative traffic accident crash data for five years from 2015 to 2019 is researched upon and considered. In the process of analysis firstly hotspot analysis is carried out using Getis-Ord Gi* method in QGIS package and then followed with Hotspot Analysis using NNH clustering in CrimeStat software to test the statistical level of significance for hotspot locations. Based on the density estimate the statistically significant hotspots are ranked. The above research methodology can be demonstrated to any size of road network area globally having relevant traffic accident data for hotspot identification and ranking for reducing the traffic accidents. Keywords - Traffic Accidents, Hotspots, Getis-Ord Gi*, Nearest Neighborhood Clustering, Spatial Statistics. | |||||||||
View Paper |